Natural History of Storming in Severe Adolescent Traumatic Brain Injury (TBI): An Observational and Descriptive Study

By: 

Lisa Cantore CPNP-AC; Julia Buch Wamstad Psy D; Ken Norwood MD; James Blackman MD, MPH; Peter D. Patrick PhD., MS

An adolescent is admitted to a pediatric rehabilitation center for further management after sustaining a traumatic brain injury. On admission, the patient is diaphoretic, posturing, febrile and hypertensive.  After stimulation is decreased and medication is administered, the patient is diagnosed with “storming.” Case studies have reported these findings in intensive care units as well as rehabilitation centers following brain injury. “Storming” has been reported in males and females, both adult and pediatric patients [1, 2, 3, and 4]. Younger age and diffuse axonal injury have been considered risk factors for the development of storming [5].

Dating back to 1929, multiple terminologies have been used to describe this phenomenon in the literature including dysautonomia, storming, paroxysmal autonomic instability with dystonia, and paroxysmal sympathetic hyperactivity [6]. It is estimated that the incidence is ranging from 15-33% following all brain injury [7]. However, a unified nomenclature has not been accepted, making this clinical entity difficult to study and hard to track clinically as a medical co-morbidity.

The pathophysiology of “storming” is also under debate. Although the original theory suggested an epileptogenic source, other case studies suggest that seizure activity is not the cause [8]. A disconnection theory has also been hypothesized [9]. This theory states that there is a functional disconnection between the brain and the spinal cord, with most authors speculating that the phenomenon could be related to autonomic dysfunction. The proposed sites of dysfunction have varied from the brainstem, diencephalon, and hypothalamus, to cortical and sub cortical centers. An excitatory: inhibitory ratio (EIR) model has also been proposed, suggesting that paroxysms are triggered by abnormal processing of stimuli within the spinal cord. One advantage of the EIR theory over the conventional disconnection theory includes the ability to explain the cluster of symptoms related to the over-reactivity associated with stimuli [8].

There are multiple clinical features that have been reported, either separately or in combination. Features include hyperthermia, hypertension, tachycardia, tachypnea, diaphoresis, agitation, and posturing [10, 11, 12, 13]. Vital sign parameters have been proposed to help define the syndrome, although some believe that the criteria have been too restrictive [1]. It has also been noted that many patients have displayed only one or two of the symptoms during an episode instead of all of them [3].

These clinical manifestations can be signs of life threatening conditions, such as sepsis or dehydration. Such conditions must be excluded prior to labeling these signs as storming [3]. However, the diagnosis is currently based on clinical observation of the specific cluster of signs (tachypnea, tachycardia, hypertension, hyperthermia, decerebrate or decorticate posturing, increased muscle tone, and profuse sweating) after excluding other possibilities [11]. Recognition and diagnosis is necessary to reduce unnecessary and additional diagnostic testing.

Pharmacological interventions are often used to stabilize vital signs and reduce occurrences, with the goal being to maximize effectiveness while minimizing adverse effects. Multiple agents have been investigated and used for treatment including opiate agonist, GABA A agonist, GABA B agonist, alpha antagonist, beta antagonist, dopamine agonist, dopamine antagonist and agonist [12, 14, 15, 16, 17]. Various case studies have reported using different agents and combinations with clinically observed benefit. Current evidence at this time does not suggest benefit of one medication regimen over another.

Methods

The University of Virginia Kluge Children’s Rehabilitation Center (KCRC) has been caring for children with brain injury for over thirty years.  A focus of interest has been children and adolescents who do not regain functional arousal or awareness following severe brain injury [18]. These children can be described as being in a prolonged state classified as Rancho Los Amigos I, II and III. A database has been established to collect ongoing differences and similarities for low response children and adolescents that have sustained a brain injury. A subset of patients in the database was diagnosed as exhibiting “storming.” In this study, we investigated the specific expression of clinical signs over time as well as whether the expression of signs was rhythmic in nature within a group of severely injured adolescents with traumatic brain injury.

We used a retrospective observational case series design that longitudinally assessed the occurrences of storming. Twelve adolescents with the diagnosis of storming who sustained a severe traumatic brain injury were included in the study (Table 1).

Table 1: Characteristics of storming patients

Number

Age

GCS at Time of Injury

Date Past Injury

Injury Type

Medications

3

19

unknown

31

TBI/MVA

labetolol

5

17

unknown

40

TBI/MVA

 

7

17

5

12

TBI/MVA

clonidine

propanolol

11

21

4

31

TBI Pedestrian versus car

metoprolol

13

11

3

30

TBI Bike accident

metoprolol

16

10

5

17

TBI Go-cart accident

 

17

20

unknown

37

TBI assault

clonidine

metoprolol

18

16

unknown

30

TBI/MVA

clonidine

19

17

5

29

TBI/MVA

labetolol

20

16

unknown

25

TBI/MVA

atenolol

24

18

4

35

TBI/MVA

clonidine,

25

9

5

16

TBI/MVA

 

Each child was monitored for a minimum of four consecutive days as an inpatient at KCRC. The range of days monitored was four to fourteen. Each twenty-four hour period was broken into four equal portions and variables recorded included blood pressure, temperature, respiratory rate, and heart rate. The Department of Pediatrics at the University of Virginia has established and developed an institutional set of vital signs parameters. Using established University of Virginia's Children’s Hospital's clinically significant cut off scores for age, the occurrences were tabulated for each child when one or more of the variables were in the clinically significant range (Table 2). Dystonia or posturing assessments were not available for analysis.

Table 2: Vital signs parameters for age

Ages 15-21

 

Pulse

Greater than or equal to 130 BPM

Respirations

Greater than or equal to 25 per minute

Systolic Blood  Pressure

Greater than or equal to 140 MMHG

Diastolic Blood Pressure

Greater than or equal to 95 MMHG

Temperature

Greater than or equal 38 degrees Celsius

Ages 7-12

 

Pulse

Greater than or equal to 135 BPM

Respirations

Greater than or equal to 30 per minute

Systolic Blood Pressure

Greater than or equal to 230 MMHG

Diastolic Blood Pressure

Greater than or equal to 90 MMHG

Temperature

Greater than or equal to 38 degrees Celsius

Data analysis consisted of tabulating a frequency count for the number of times each variable met clinical threshold for blood pressure, respiration, temperature, and heart rate during each of four, six hour time blocks (12am-6am, 6am-12pm, 12pm-6pm, and 6pm-12am). Each variable was plotted on the child’s graph. A descriptive analysis was used to characterize the group as well as individual tendencies in an attempt to identify a rhythmic pattern of occurrence (Figure 1). Each analysis yielded an individual graph of response for days mon itored.

Figure #1:  16 Year Old Male 30 Days Post Injury

(Vertical Axis # of Episodes, Horizontal Axis Time Periods)

Fig #2  20 Year Old Male 39 Days Post Injury

Fig #3  17 Year Old Male 17 Days Post Injury

Results

Descriptive analysis of each child’s “natural history” of events leads to multiple conclusions.

  1. At no time did all four variables elevate into the clinically significance range for any of the 12 children.
  2. Typically one or two variables characterized the adolescent’s event graph.
  3. There was no group rhythmic pattern detected.
  4. Individuals demonstrated rhythmic patterns on their event graphs with a tendency toward one or two variable characterization.  For example one child may have clinically elevated heart rate or temperature at specific time episodes.  In contrast, another child may have different or similar variable elevations at a differing time.
  5. Finally, in spite of having the diagnosis of “storming,” once monitored several children never met clinically significant threshold for any of the four variables monitored.  In practical terms, while the child was described as “storming” they in fact did not meet clinical criteria for this study.  These points to the importance of using objective clinical data to make a diagnosis of a “storming” episode as opposed to relying only on bedside observation.

Conclusions

Individual patterns with a predominance of variable expression can inform medical decision making regarding “storming” and inform management decision making including influence medication selection. While medical “rule outs” of threatening medical conditions must always be the initial response to symptoms of “storming”, increased knowledge of natural history and group as well as individual patterns of expression helps “storming” management.  Using a bedside descriptive diagnostic model may be better served by a monitoring and measurement program over sufficiently long enough time periods to empirically support a “storming” diagnosis.   In addition, the diagnosis of “storming” may better serve decision making when stated, “Storming with a predominance of tachycardia, or hypertension or hyperthermia or tachypnea during morning, afternoon, or night time episodes”. This way of characterizing the adolescent’s “storming” better serves the attending medical team’s intervention decisions.  For example, medication selection and timing can be individualized and focused on managing the predominant clinical manifestation as it pertains to the adolescent’s event expressions.   Also, measures of improvement/outcome can be more specifically focused for each adolescent in the medical team’s efforts to measure effectiveness and improvement.

In the absence of randomized controlled trials and higher quality levels of evidence, efforts must continue to gather practice-based evidence to inform and advance medical team decision making of “storming".  Future efforts to study various age groups and to use prospective methodological investigations must be developed and supported.

References:

  1. Rabinstein, A. (2007). Paroxysmal Sympathetic Hyperactivity in the Neurological Intensive Care Unit.Neurological Research, 29 (7): 680-682.
  2. Tanti, A., Gasperini, G., & Rossini, M. (2005). Paroxysmal Episodic Hypothalamic Instability with Hypothermia after Traumatic Brain Injury. Brain Injury, 19 (14): 1277-1283.
  3. Blackman, J.A., Patrick P.D., Buck, M.L., & Rust, R.S. (2004). Paroxysmal Autonomic Instability with Dystonia after brain injury. Archives of Neurology, 61:321-328.
  4. Srinivasan S., Lim, C., & Thirugnanam, U. (2007).  Paroxysmal Autonomic Instability with Dystonia. Clinical Autonomic Research, 17 (6).
  5. Li-Quan, L., Li-Jun, H., Ming-Kun, Y. et al (2011). Risk Factors Related to Dysautonomia After Severe Traumatic Brain Injury. Journal of Trauma-Injury Infection & Critical Care, 71: 538-542.
  6. Baguley, I. (2008). Autonomic Complications Following Central Nervous System Injury. Semin Neurology. 28: 716-725.
  7. Baguley, I., Slewa-Younan, S., Heriseanu, R., Nott, M., Mudaliar, Y., & Nayyar, V. (2007). The Incidence of Dysautonomia and its Relationship with Autonomic Arousal Following Traumatic Brain Injury. Brain Injury, 21 (11): 1175-1181.
  8. Baguley, I., Heriseanu, R., Cameron, I., Nott, M., & Slewa-Younan, S. (2007). A Critical Review of the Pathophysiology of Dysautonomia Following Traumatic Brain Injury. Neurocritical Care, 8(2): 293-300.
  9. Boeve, B., Wijdicks, E., Benarroch, & Schmidt. (1998). Paroxysmal Sympathetic Storms (Diencephalic Seizures) after Severe Diffuse Anxonal Head Injury. Mayo Clinic Proceedings, 73 (2): 148-152.
  10. Pranzatelli, M. (1991). Hypothalamic-Midbrain Dysregulation Syndrome: Hypertension, Hyperthermia, Hyperventilation and Decerebration. Journal of Child Neurology, 6 (2): 115-122.
  11. Baguley, I., Nicholls, J., Felmingham, K., Crooks, J., Gurka, J., & Wade, L. (1999). Dysautonomia after Traumatic Brain Injury: a Forgotten Syndrome? J Neurol Neurosurg Psychiatry 67: 39-43.
  12. Baguley, I,. Nott, M., Slewa-Younan, S., Heriseanu, R., & Perkes, I. (2009). Diagnosing Dysautonomia After Acute Traumatic Brain Injury: Evidence for Overresponsiveness to Afferent Stimuli. Arch Physical Med Rehabilitation 90: 580-586.
  13. Kishner, S., Augustin, J., & Strum, S. (2008). Post Head Injury Autonomic Complications. Physical Medicine and Rehabilitation.
  14. Wang,V., and Manley, G. (2007). Recognition of Paroxysmal Autonomic Instability with Dystonia (PAID) in a Patient with Traumatic Brain Injury. The Journal of Trauma: Injury, Infection and Critical Care, 64 (2): 500-502.
  15. Rossitch, E., & Bullard, D. (1988). The Autonomic Dysfuction Syndrome: Aetiology and treatment. British Journal of Neurosurgery, 2 (94): 471-478.
  16. Hendricks, H., Heeren, A., & Vos, P. (2010). Dysautonomia After Severe Traumatic Brain Injury. European Journal of Neurology, 17: 1172-1177.
  17. Walker, M., & Holloway, R. (2006). Storming: Transient Autonomic Dysfunction in Persistent Vegetative and Minimally Conscious State. Journal of Hospice and Palliative Nursing, 8(4): 198-199.
  18. Patrick, P.D., Blackman, J.A., Mabry, J.L., Buck, M.L., Gurka, M.J., & Conaway, M.R. (2006) Dopamine agonist therapy in low-response children following traumatic brain injury. Journal of Child Neurology, 21(10):879-885.